Authors:
Antonio Carlos Meira Neto
;
Rafael Gaspar de Sousa
;
Marcelo Fantinato
and
Sarajane Peres
Affiliation:
School of Arts, Science and Humanities, University of São Paulo, São Paulo, Brazil
Keyword(s):
Concept Drift, Process Mining, Transition Matrix, Event Log, Process Drift, Data Mining.
Abstract:
Contemporary process mining techniques commonly assume business processes are in a steady state. However, business processes are prone to change and evolution in response to various factors, which can happen at any time, in a planned or unplanned way. This phenomenon of business process evolution and change is known as concept drift, and identifying and understanding is of paramount relevance for business process management, so that organizations can respond and adapt to the new challenges they face. The goal of this paper is to introduce the use of transformed transition matrices as a data structure to support the treatment of concept drifts in process mining, given its efficiency, simplicity, and expandability. The proposed data structure allows to handle different concept drift aspects in an integrated way. Three concept drift detection methods are first adapted to work on transformed transition matrices. The results obtained in the experiments are compared with a state-of-the-art
method (baseline), and the three methods used achieved good results, showing an encouraging potential for future planned work.
(More)